RESEARCH ARTICLE Evaluation of Rint1 as a modifier of intestinal tumorigenesis and cancer risk Karla L. Otterpohl¤, Karen A. Gould* Department of Genetics, Cell Biology and Anatomy, University of Nebraska Medical Center, Omaha, Nebraska, United States of America ¤ Current address: Children's Health Research Center, Sanford Research, Sioux Falls, South Dakota, United States of America * [email protected] a1111111111 Abstract a1111111111 a1111111111 The Rad50 Interacting Protein 1 (Rint1) influences cellular homeostasis through mainte- a1111111111 nance of endoplasmic reticulum, Golgi and centrosome integrity and regulation of vesicle a1111111111 transport, autophagy and the G2/M checkpoint. Rint1 has been postulated to function as a tumor suppressor as well as an oncogene, with its role depending perhaps upon the precise cellular and/or experimental context. In humans, heterozygosity for germline missense vari- ants in RINT1 have, in some studies, been associated with increased risk of both breast and OPEN ACCESS Lynch syndrome type cancers. However, it is not known if these germline variants represent Citation: Otterpohl KL, Gould KA (2017) Evaluation loss of function alleles or gain of function alleles. Based upon these findings, as well as our of Rint1 as a modifier of intestinal tumorigenesis initial consideration of Rint1 as a potential candidate for Mom5, a genetic modifier of intesti- and cancer risk. PLoS ONE 12(3): e0172247. Min/+ doi:10.1371/journal.pone.0172247 nal tumorigenesis in Apc mice, we sought to explicitly examine the impact of Rint1 on tumorigenesis in ApcMin/+ mice. However, heterozygosity for a knockout of Rint1 had no Editor: Alvaro Galli, CNR, ITALY impact on tumorigenesis in Rint1+/-; ApcMin/+ mice. Likewise, we found no evidence to sug- Received: November 15, 2016 gest that the remaining Rint1 allele was lost somatically in intestinal tumors in ApcMin/+ mice. Accepted: February 1, 2017 Interestingly, in contrast to what has been observed in Rint1+/- mice on a mixed genetic +/- Published: March 6, 2017 background, Rint1 mice on a pure C57BL/6J background did not show spontaneous Copyright: © 2017 Otterpohl, Gould. This is an tumor development. We also evaluated colorectal cancer data available in the COSMIC and open access article distributed under the terms of ONCOMINE databases and found that RINT1 overexpression, as well as the presence of the Creative Commons Attribution License, which somatic missense mutations in RINT1 were associated with colorectal cancer development. permits unrestricted use, distribution, and In vitro evaluation of two missense variants in RINT1 suggested that such variants do have reproduction in any medium, provided the original author and source are credited. the potential to impact RINT1 function. Data Availability Statement: All relevant data are within the paper. Funding: The University of Nebraska DNA Sequencing Core receives partial support from the Introduction NCRR (5P20RR016469, RR018788) and the National Institute for General Medical Science The Rad50 interacting protein 1 (Rint1), was identified through its interaction with Rad50, (NIGMS) (8P20GM103427, P20GM103471). The and found to be involved in control of the G2/M checkpoint in response to DNA damage [1]. author(s) received no other specific funding for this Depletion of Rint1 in vitro results in genomic instability, failure to complete the G2/M check- work. point and ultimately cell death [2, 3]. Rint1 has also been shown to function in a variety of Competing interests: The authors have declared other cellular processes, including telomerase-independent telomere elongation, centrosome that no competing interests exist. duplication, endoplasmic reticulum (ER) and Golgi homeostasis, ER-Golgi vesicle trafficking, PLOS ONE | DOI:10.1371/journal.pone.0172247 March 6, 2017 1 / 13 Rint1 as a modifier of tumorigenesis ER stress and autophagy[3±8]. Homozygosity for a Rint1 knockout allele in mice results in embryonic lethality during early postimplantation development, apparently due to a failure of the blastocysts to proliferate [2]. However, selective deletion of Rint1 in non-dividing Purkinje cells also results in cell death [3]. However, in this later case, cell death appeared to be the result of defects in ER-Golgi function and inhibition of autophagy [3]. Several studies have provided evidence to suggest that heterozygosity for mutations in the Rint1 gene is associated with tumorigenesis. Approximately 80% of mice heterozygous for the Rint1 knockout allele (Rint1 tm1Whl/+) were reported to develop spontaneous lymphomas or solid tumors affecting a variety of tissues including the liver, lung, and uterus [2]. However, the average life span of these mice was 2 years, indicating that tumors developed with a long latency period. This study suggests that Rint1 may function as a general tumor suppressor gene. By contrast, another study identified Rint1 as a potential oncogene in glioblastoma [9]. Over-expression of Rint1 promoted anchorage-independent growth of glioblastoma cell lines, whereas shRNA knockdown of Rint1 significantly reduced cell viability. Most recently, germline variants in Rint1 have been associated with increased risk for breast can- cer as well as Lynch syndrome type cancers, including colorectal, endometrial, and gastric cancers [10]. Whole exome sequencing in families with previously unexplained hereditary breast cancer identified three different germline Rint1 variants, each found in a single fam- ily [10]. Focused analysis subsequently identified more than 20 different rare germline vari- ants in RINT1 predicted to be damaging that were associated with breast cancer in a case control study and 5 additional predicted damaging germline RINT1 variants associated with breast, colorectal, endometrial and gastric cancer in six cancer±prone families [10]. Inter- estingly, the vast majority of the predicted damaging germline variants in RINT1 that were associated with increased risk of breast and Lynch syndrome type cancers in this study were missense variants. The potential impact of these variants on Rint1 function is unknown. However, it should be noted that a recent study found no evidence for an association between germline missense variants in RINT1 and either breast or Lynch syndrome type cancer risk [11]. Our laboratory initially identified Rint1 as a potential candidate for the Mom5 modifier of intestinal neoplasia in ApcMin/+ mice [12]. The B6 allele of the Mom5 locus is associated with increased intestinal adenoma number and size, and the B6 allele of Rint1 harbors two missense mutations that are similar to some of the RINT1 variants linked to breast and Lynch syndrome type cancers in humans. Both variants in the B6 allele result in the substitution of a highly con- served isoleucine residue in the coiled-coil domain of Rint1 [12]and thus have the potential to alter the folding of the coiled-coil of Rint1, the ability of Rint1 to interact with its binding part- ners and/or alter the stability of the Rint1 protein. Rint1 interacts with several proteins via its coiled-coil domain, including Zw10, Uvrag, Cog1, and Stx16 [6, 7, 13]. All these proteins are involved in membrane trafficking between the ER and Golgi. Although fine structure mapping ultimately excluded Rint1 as a candidate for Mom5 [14], because RINT1 mutations had been associated with increased risk of Lynch syndrome type cancers in humans, we were interested in explicitly examining the impact of Rint1 mutation on intestinal tumorigenesis in ApcMin/+ mice. Although these in vivo studies suggested that heterozygosity for a knockout of Rint1 did not enhance intestinal tumorigenesis in ApcMin/+ mice, in vitro studies indicated that missense mutations in Rint1 do have the potential to impact protein-protein interactions involving Rint1. We also found that over-expression of RINT1 and somatic missense mutations in RINT1 were relatively frequent in human colorectal cancers whereas loss of RINT1 or RINT1 expression were infrequent events in these cancers. PLOS ONE | DOI:10.1371/journal.pone.0172247 March 6, 2017 2 / 13 Rint1 as a modifier of tumorigenesis Materials and methods Care and production of B6 mice congenic for the Rint1 knockout allele All mice used were housed in facilities at the University of Nebraska Medical Center accredited by the American Association for Accreditation of Laboratory Animal Care. Mice were kept in a climate controlled environment with 14-hour light/10-hour dark cycles and access to Harlan irradiated rodent diet 7904 (Harlan Teklad, Madison, WI) and water ad libitum. All proce- dures involving live mice were approved by the University of Nebraska Medical Center Institu- tional Animal Care and Use Committee. Four Rint1tm1Whl/+ (Rint1+/-) mice, heterozygous for a targeted disruption of exons 5 and 6 of Rint1 (Lin et al. 2007), were obtained from Dr. Wen Hwa Lee (University of California- Irvine, Irvine, California). These mice were on a mixed 129SvEv (129) and C57BL/6J (B6) genetic background. Thus, the genotype at SNPs distributed throughout the genome was determined at the DartMouse™ Speed Congenic Core Facility at the Geisel School of Medicine at Dartmouth University using the Illumina, Inc. (San Diego, CA) GoldenGate Genotyping Assay. The raw SNP data were analyzed using SNaP-Map™ and Map-Synth™ software. Analysis of the genetic background of the founder mice revealed that all mice carried residual 129 alleles spanning a 32 Mbp region linked to the Rint1 knockout allele (Fig 1A). Additionally, these